Establishment of a 3.83-Ga magmatic age for the Akilia tonalite (southern West Greenland)

Establishment of a 3.83-Ga magmatic age for the Akilia tonalite (southern West Greenland)

Earth and Planetary Science Letters 202 (2002) 563^576 www.elsevier.com/locate/epsl Establishment of a 3.83-Ga magmatic age for the Akilia tonalite (...

435KB Sizes 27 Downloads 74 Views

Earth and Planetary Science Letters 202 (2002) 563^576 www.elsevier.com/locate/epsl

Establishment of a 3.83-Ga magmatic age for the Akilia tonalite (southern West Greenland) Stephen J. Mojzsis  , T. Mark Harrison 1 Department of Earth and Space Sciences and Institute of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90095-1567, USA Received 1 February 2002; received in revised form 17 May 2002; accepted 5 July 2002 This paper is dedicated to the late Vic McGregor who ¢rst recognized, described, and mapped the rocks of the ‘Akilia association’ in southern West Greenland.

Abstract A recurring challenge to ion microprobe U^Pb zircon geochronology has been to discriminate between preservation of original igneous zircon populations and inherited grains. This has proved particularly problematic in studying the polyphase metamorphic rocks that dominate Early Archean gneissic terranes. For example, differing interpretations exist for the origin of complex zircon populations investigated by U^Pb ion microprobe zircon geochronology from a granitoid (orthogneiss) body, previously used to establish a minimum age for a supracrustal enclave on Akilia (island) in southern West Greenland. We describe a method whereby the geochemistry of the Akilia orthogneiss, coupled with a U^Th^Pb vs. age depth profile in a zircon from the same rock, permits direct assessment of zircon inheritance. Results reveal evidence for three phases of concentric zircon growth at 3.83 Ga, V3.6 Ga and 2.7^2.5 Ga; zircon growth at both V3.6 Ga and 2.7^2.5 Ga is consistent with precipitation from a metamorphic fluid. Depth profile U^Th^Pb data demonstrate that only the s 3.8-Ga zircon core could have crystallized from the host rock. We conclude that the magmatic age of this rock is 3.83 7 0.01 Ga, not V3.65 Ga as has been previously proposed. The U^Th^Pb zircon depth profile technique has wider applications to resolving other geochronological debates in highgrade metamorphic terranes where zircons populations are diverse and individual grains record complex overgrowth and dissolution/re-precipitation features. : 2002 Elsevier Science B.V. All rights reserved. Keywords: Archean; West Greenland; zircon; geochronology; life origin; ion probe data

1. Introduction * Corresponding author. Present address: Department of Geological Sciences, UCB 399, University of Colorado, Boulder, CO 80309-0399, USA. Tel.: +1-303-492-5014; Fax: +1-303-492-5014. E-mail address: [email protected] (S.J. Mojzsis). 1 Present address: Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia.

Early Archean supracrustal rocks, preserved within highly metamorphosed granitoid gneiss complexes, provide the only direct record of the state of the continental crust, hydrosphere and biosphere prior to V3.5 Ga. These ancient volcano-sedimentary assemblages form the baseline for understanding early planetary processes, including the timing of the emergence of life on Earth and

0012-821X / 02 / $ ^ see front matter : 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 1 2 - 8 2 1 X ( 0 2 ) 0 0 8 2 5 - 7

EPSL 6344 11-9-02

564

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

its potential to have appeared elsewhere in the solar system. In these terranes, sediments typically appear as highly metamorphosed and strongly deformed supracrustal enclaves, ranging in size from less than a meter wide to tens of kilometers in length. All are locked within voluminous granitoid gneisses [1]. The oldest recognized supracrustal enclave [2,3] is a V5-m-thick layer of quartz^magnetite ‘ironstone’ interposed between 10^40-mthick layers of amphibolite and ultrama¢c rocks on the southwestern point of Akilia (island) [4], Ameralik fjord region, West Greenland (Fig. 1). Tonalitic to granodioritic orthogneisses on Akilia found in contact with the northern and eastern margins of the supracrustal enclave yield complex U^Pb zircon ages as old as 3.87 Ga [3,5^7]. These results have been interpreted [3,5] as the age of the igneous protoliths to the orthogneisses and provide a minimum age of the associated supracrustal rocks. Carbon isotope data [2] have been interpreted to indicate bioorganic activity during deposition of the metasedimentary rocks on Akilia (cf. [35]). If these interpretations are correct, it would mean that life emerged on Earth signi¢cantly earlier than previously established [2,3,8^10]. Other carbon isotope evidence from the Isua supracrustal belt 150 km northeast of Akilia, places the emergence of life at least before 3.77 Ga [8,11], thereby raising the likelihood for an early and rapid emergence of the biosphere on Earth. However, several authors [12,13] have argued based on whole-rock and mineral common Pb data that the geologically diverse ‘Am|“tsoq’ orthogneisses of West Greenland, including those that host Akilia sedimentary enclaves, are the product of a single V3.6-Ga crustal formational event. This interpretation is in con£ict with the view that the at present large (3000 km2 ) and geologically heterogeneous ‘Itsaq Gneiss Complex’ [5] formed in a protracted fashion from V3.9 to 3.6 Ga. If the age of generically termed ‘Am|“tsoq’ granitoid crystallization is actually ca. 3.6 Ga, this would suggest that all s 3.8-Ga zircons contained in various orthogneisses of West Greenland thus far analyzed, including those on Akilia and nearby islands, are inherited. Resolving the question of inheritance vs. protolith igneous origin of complex zircons, particu-

larly with regard to the age of the orthogneisses on Akilia, has implications for the age of rocks hosting the oldest known sediments and reported evidence for life contained therein [2], as well as for the timing and nature of events at the surface during the emergence of the biosphere. Con¢rmation of a s 3.8-Ga age for the Akilia orthogneisses and the associated metasediments and their biological signatures is signi¢cant because it would place their deposition prior to termination of an extended (and ill-de¢ned) period of meteoritic and cometary bombardment of the inner solar system centered at ca. 3.85 Ga [14,15] and coeval with the presence of liquid water on early Mars [16,17].

2. Zircon growth events in the Archean (V V3.8 Ga, 3.6 Ga and 2.7 Ga) There is general agreement that the gneisses on Akilia contain at least three zircon age populations: ca. 3.85 Ga, V3.6 Ga, and V2.7 Ga [3,5^7]. A recently proposed explanation [6,7,13] for this age variation is that the V2.7-Ga zircons are metamorphic overgrowths, the V3.6-Ga grains are igneous and all of the oldest grains are inherited [12]. That the tonalitic protolith of the Akilia orthogneiss in question can only be V3.6 Ga [6,7,12,13] is problematic in our view for the following reasons. (1) The strongly zircon undersaturated [18] nature of the original magma (Table 1 ; Fig. 2) intruding at V950‡C [19] into zircon-poor supracrustal rocks such as amphibolite and hornblendite on Akilia is unlikely to preserve widespread inherited zircon. Tonalitic orthogneiss sample GR9716 [16] studied here (equivalent to G93/05 of Nutman et al. [3,5] and SM/GR/97/7 of Whitehouse et al. [6,7]) has a Zr abundance of 144 ppm (Table 1) and the observed whole rock composition is probably close to that of the protolith magma [19]. Thus, at temperatures V950‡C and M = (Na+K+2Ca/SiUAl) [18] of 1.54, the solubility of zircon in the tonalite melt would have been close to 900 ppm (Fig. 2) far below the 144 ppm Zr present in GR9716.

EPSL 6344 11-9-02

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

565

Fig. 1. Locality map showing Akilia (island) within the Early Archean ‘Itsaq Gneiss Complex’ in the vicinity of Ameralik (fjord) in southern West Greenland (inset). A complex association of supracrustal rocks and associated gneisses are found on Akilia (island). Adjacent islands and coastal areas contain similar relationships [3^5,22,26,27] that await further detailed study. Map modi¢ed after A.P. Nutman (unpublished manuscript).

EPSL 6344 11-9-02

566

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

(2) While Kamber and Moorbath [13] emphasized the lack of 3.8-Ga common Pb in feldspars from Akilia orthogneisses, metamorphism at V3.6 Ga occurred under granulite grade conditions ( s 750‡C and 7 kbar; [20]) that would readily cause exchange of primitive Pb isotopes in feldspar [21] with radiogenic Pb released from U-rich phases, dominantly zircon. Thus, their observation [13] is the expected result of a 3.6-Ga high-grade metamorphic event and does not preclude a 3.8-Ga protolith age. It has been pointed out [22] that the diverse rocks arbitrarily assigned a common ‘Am|“tsoq gneiss’ origin and analyzed together by Kamber and Moorbath [13] are in fact petrogenetically unrelated and therefore do not provide data relevant to the ages of the Akilia supracrustal enclaves or the orthogneisses that surround and in some cases, intrude them. Hence, the controversy we address here is based on zircon U^Pb isotope geochronology on an Akilia orthogneiss that was originally assigned an igneous protolith age of 3.85 Ga [3] but was subsequently reevaluated by Whitehouse et al. [6,7] who have argued for widespread zircon inheritance and a protolith age of V3.65 Ga. Table 1 Whole rock geochemistry of sample GR9716 Major elements SiO2 Al2 O3 Fe2 O3 MnO MgO CaO Na2 O K2 O TiO2 P2 O5 LOI Total

wt% 65.5 16.9 3.91 0.06 2.26 4.65 3.75 1.31 0.38 0.29 0.54 99.6

Trace elements

ppm

Zr Th U Pb

144 1.8 0.5 8.0

M = 1.54a a

[18].

Fig. 2. Zircon saturation curves [18] plotted for a source temperature of 950‡C and lower temperatures (800 ‡C, 775 ‡C) plotted near the freezing point of tonalitic melts [19]. We note that at the source temperature of 950‡C all homogeneous orthogneisses identi¢ed from southern West Greenland that provide U^Pb zircon ages older than 3.8 Ga [4,5,16,22] are severely undersaturated in Zr and any zircon entrained in such melts would have a strong likelihood of dissolving. Hence, the chance for widespread zircon inheritance (cf. [6,7,12,13]) in any of these rocks is low.

In this paper, we undertake two new styles of analysis in order to evaluate the likelihood of zircon inheritance in the Akilia tonalitic orthogneisses, i.e. U^Th^Pb isotope depth pro¢ling of individual zircons and in situ analysis of zircon in petrographic thin section.

3. Methods We analyzed zircons from the same tonalitic orthogneiss locality collected from the northern margin of the southwestern peninsula of Akilia, West Greenland (Fig. 1), from which Nutman et al. [3], Whitehouse et al. [6,7] and Mojzsis and Harrison [16] (sample GR9716) obtained their results. 3.1. Conventional ion microprobe U^Pb geochronology All U^Pb ion microprobe analyses of zircons, extracted by conventional heavy mineral separa-

EPSL 6344 11-9-02

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

Fig. 3. Compilation of 207 Pb/206 Pb ion microprobe ages of zircons from orthogneiss sample GR9716 in contact with the supracrustal enclave on Akilia [3,5^7,22,27]. The zircon population at V3.85 Ga, interpreted here to represent the magmatic age of the granitoid, has uniformly high Th/U values. Younger zircons generally exhibit lower Th/U values more characteristic of a metamorphic origin and correlate with well-documented metamorphic events in West Greenland at V3.6 Ga and V2.7 Ga [3^5,20,22].

567

tion of whole-rock powders or by in situ measurements in petrographic thin section, were determined using the UCLA CAMECA ims 1270 ion microprobe. Separated zircon grains were handpicked, mounted in epoxy and polished on 0.25 Wm alumina ; all samples were coated with W of Au using a sputter-coater. The stanV100 A dard operating conditions for analysis presented primary beam focused to a here are a O3 2 V20U25 Wm spot; the ion microprobe was operated at a mass resolving power of 6000 with an energy window of 50 eV. An o¡set voltage of 5^ 15 eV was used for 238 Uþ relative to Pbþ and UOþ to compensate for their contrasting energy distributions. Oxygen £ooding to a pressure of 3.5U1035 torr was employed to increase Pbþ yields. U^Pb ages were determined by comparison with a working curve de¢ned by multiple measurements of standard zircon AS-3 that yield concordant 206 Pb/238 U and 207 Pb/235 U ages of 1099.1 7 0.5 Ma [23]. All studies to date of the contested orthogneiss outcrop on Akilia using U^Pb zircon spot ages on separated zircons yield a similar age distribution and a general trend toward lower Th/U ratio with

Fig. 4. Optical micrograph of unpolished euhedral zircon grain GR9716_69 used in U^Th^Pb depth pro¢le analysis.

EPSL 6344 11-9-02

568

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

Table 2 Ion microprobe depth pro¢le analyses of zircon GR9716_69 Depth (Wm) 0.00 0.19 0.37 0.56 0.74 0.93 1.12 1.30 1.49 1.67 1.86 2.05 2.23 2.42 2.60 2.79 2.98 3.16 3.35 3.53 3.72 3.91 4.09 4.28 4.47 4.65 4.84 5.02 5.21 5.40 5.58 5.77 5.95 6.14 6.33 6.51 6.70 6.88 7.07 7.26 7.44 7.63 7.81 8.00 8.19 8.37 8.56 8.74 8.93 9.12 9.30 9.49 9.67

206

Pb*/238 U

0.455 7 0.011 0.471 7 0.011 0.479 7 0.016 0.536 7 0.012 0.574 7 0.025 0.582 7 0.033 0.550 7 0.011 0.546 7 0.013 0.567 7 0.015 0.551 7 0.011 0.537 7 0.018 0.522 7 0.011 0.439 7 0.011 0.415 7 0.015 0.358 7 0.011 0.325 7 0.009 0.310 7 0.006 0.299 7 0.005 0.515 7 0.028 0.538 7 0.011 0.559 7 0.017 0.717 7 0.134 0.786 7 0.094 0.705 7 0.153 0.823 7 0.069 0.782 7 0.069 0.800 7 0.120 0.638 7 0.094 0.714 7 0.229 0.600 7 0.066 0.806 7 0.105 0.850 7 0.136 0.717 7 0.082 0.782 7 0.059 0.838 7 0.059 0.840 7 0.107 0.772 7 0.082 0.848 7 0.063 0.654 7 0.074 0.607 7 0.040 0.745 7 0.095 0.730 7 0.108 0.885 7 0.177 0.717 7 0.074 0.846 7 0.126 0.759 7 0.099 0.818 7 0.111 0.801 7 0.116 0.626 7 0.074 0.778 7 0.090 0.818 7 0.182 0.744 7 0.143 0.708 7 0.168

207

Pb*/235 U

10.96 7 0.35 11.46 7 0.26 12.03 7 0.37 12.33 7 0.31 13.62 7 0.55 14.27 7 0.79 13.34 7 0.25 13.54 7 0.31 14.10 7 0.34 14.02 7 0.29 13.79 7 0.50 13.31 7 0.30 11.30 7 0.27 10.54 7 0.34 9.15 7 0.24 8.33 7 0.24 8.00 7 0.16 7.69 7 0.13 12.01 7 0.69 13.10 7 0.29 13.73 7 0.40 30.70 7 5.57 33.83 7 3.98 30.80 7 6.49 37.66 7 3.10 36.11 7 3.19 36.20 7 5.42 32.06 7 5.07 36.24 7 11.71 30.31 7 3.33 35.79 7 4.57 38.56 7 6.30 31.45 7 3.64 35.05 7 2.64 37.29 7 2.64 36.97 7 4.99 34.97 7 3.41 39.03 7 2.86 29.81 7 3.55 28.13 7 1.92 34.89 7 4.49 35.06 7 5.49 43.50 7 8.76 35.81 7 3.70 43.17 7 6.48 38.17 7 5.17 41.72 7 5.38 40.85 7 5.99 32.08 7 3.95 39.40 7 4.26 41.04 7 9.25 37.45 7 7.14 36.74 7 8.63

207

Pb*/206 Pb*

0.1745 7 0.0032 0.1764 7 0.0020 0.1823 7 0.0035 0.1669 7 0.0022 0.1721 7 0.0022 0.1777 7 0.0029 0.1759 7 0.0016 0.1800 7 0.0016 0.1806 7 0.0013 0.1846 7 0.0015 0.1861 7 0.0024 0.1850 7 0.0029 0.1868 7 0.0015 0.1843 7 0.0013 0.1855 7 0.0030 0.1859 7 0.0031 0.1871 7 0.0032 0.1864 7 0.0018 0.1690 7 0.0028 0.1766 7 0.0017 0.1782 7 0.0023 0.3104 7 0.0088 0.3120 7 0.0039 0.3170 7 0.0079 0.3320 7 0.0045 0.3348 7 0.0049 0.3281 7 0.0061 0.3647 7 0.0095 0.3680 7 0.0151 0.3666 7 0.0090 0.3218 7 0.0039 0.3291 7 0.0079 0.3182 7 0.0045 0.3250 7 0.0054 0.3227 7 0.0046 0.3194 7 0.0067 0.3284 7 0.0066 0.3337 7 0.0042 0.3304 7 0.0064 0.3362 7 0.0056 0.3395 7 0.0060 0.3486 7 0.0123 0.3563 7 0.0071 0.3623 7 0.0095 0.3701 7 0.0097 0.3647 7 0.0072 0.3698 7 0.0078 0.3697 7 0.0133 0.3715 7 0.0088 0.3675 7 0.0106 0.3637 7 0.0142 0.3653 7 0.0082 0.3762 7 0.0126

207

Pb*/206 Pb* age (Ma)

%206 Pb*

Th/U

208

2602 7 17 2619 7 10 2674 7 17 2527 7 12 2578 7 12 2632 7 14 2614 7 8 2653 7 8 2658 7 7 2695 7 7 2708 7 11 2698 7 14 2714 7 7 2692 7 6 2703 7 15 2706 7 15 2717 7 15 2711 7 9 2548 7 15 2621 7 9 2636 7 11 3523 7 20 3531 7 9 3555 7 18 3626 7 9 3639 7 10 3608 7 12 3769 7 17 3783 7 27 3777 7 16 3579 7 8 3613 7 16 3561 7 9 3593 7 11 3583 7 9 3567 7 14 3610 7 13 3634 7 8 3619 7 13 3646 7 11 3660 7 12 3701 7 23 3734 7 13 3760 7 17 3792 7 17 3770 7 13 3790 7 14 3790 7 23 3798 7 15 3781 7 19 3766 7 25 3772 7 15 3816 7 20

99.78 99.73 99.89 99.65 99.93 99.98 99.95 99.97 100 99.98 99.98 100 99.99 100 99.99 99.99 100 100 99.81 99.93 99.97 99.98 99.99 99.93 100 100 99.95 100 99.86 99.83 100 99.99 99.97 100 100 100 99.96 100 100 100 99.89 99.92 99.92 100 99.95 100 100 99.8 100 99.75 100 99.75 99.87

0.0134 0.0169 0.0151 0.004 0.0047 0.0042 0.0075 0.0136 0.0147 0.0212 0.0212 0.0174 0.0226 0.0181 0.0113 0.0111 0.0084 0.0053 0.0009 0.0014 0.0012 0.0149 0.0114 0.0165 0.0162 0.0155 0.0156 0.4487 0.4135 0.4521 0.0131 0.0103 0.0118 0.0081 0.0135 0.0102 0.0166 0.0431 0.0755 0.0784 0.1111 0.1421 0.1961 0.2606 0.2754 0.3275 0.2991 0.4075 0.41 0.4266 0.3975 0.3961 0.4164

0.0093 0.0068 0.0074 0.0063 0.0029 0.0026 0.0045 0.0049 0.0055 0.0064 0.0064 0.007 0.0073 0.0061 0.0042 0.0035 0.0033 0.002 0.0054 0.0021 0.0015 0.0059 0.0048 0.0052 0.0037 0.0036 0.0037 0.1329 0.1392 0.1373 0.0039 0.0032 0.0035 0.0032 0.0029 0.0032 0.0062 0.0163 0.0223 0.0279 0.0417 0.0522 0.0687 0.095 0.1101 0.1164 0.1054 0.1182 0.1313 0.1302 0.1277 0.1305 0.1335

EPSL 6344 11-9-02

Pb*/206 Pb*

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

569

Table 2 (Continued). Depth (Wm)

206

Pb*/238 U

9.86 10.05 10.23 10.42 10.60 10.79 10.98 11.16 11.35 11.53 11.72 11.91 12.09 12.28 12.47 12.65 12.84 13.02 13.21 13.40 13.58 13.77 13.95 14.14 14.33 14.51 14.70 14.88 15.07 15.26 15.44 15.63 15.81 16.00

0.638 7 0.140 0.970 7 0.140 0.713 7 0.109 0.841 7 0.140 1.621 7 0.547 0.735 7 0.232 0.678 7 0.232 1.206 7 0.909 1.021 7 0.289 1.224 7 0.446 0.915 7 0.196 0.802 7 0.256 1.038 7 0.223 0.837 7 0.167 1.282 7 0.556 0.857 7 0.367 1.193 7 0.301 0.948 7 0.312 0.792 7 0.299 0.923 7 0.216 1.074 7 0.207 0.826 7 0.158 0.897 7 0.216 1.069 7 0.212 1.737 7 0.700 1.097 7 0.234 1.207 7 0.286 0.745 7 0.150 0.790 7 0.203 0.936 7 0.341 0.729 7 0.168 1.187 7 0.356 0.808 7 0.149 0.963 7 0.247

207

Pb*/235 U

33.09 7 7.24 50.12 7 7.60 37.81 7 5.71 43.25 7 7.23 86.47 7 28.76 38.71 7 12.30 34.49 7 12.11 61.05 7 46.28 53.21 7 14.86 66.35 7 24.01 48.08 7 10.46 41.93 7 13.07 53.96 7 11.82 43.61 7 8.81 67.58 7 29.39 43.66 7 18.61 62.12 7 15.95 49.59 7 16.01 42.50 7 15.99 49.51 7 11.63 57.28 7 10.77 44.25 7 8.36 47.01 7 11.06 57.66 7 11.47 91.77 7 37.30 56.91 7 12.26 64.15 7 14.42 38.92 7 7.84 42.19 7 10.80 49.44 7 18.23 37.86 7 8.86 62.37 7 18.54 42.20 7 8.05 51.69 7 13.16

207

Pb*/206 Pb*

0.3761 7 0.0167 0.3747 7 0.0134 0.3846 7 0.0088 0.3728 7 0.0136 0.3869 7 0.0103 0.3818 7 0.0092 0.3691 7 0.0198 0.3671 7 0.0248 0.3780 7 0.0098 0.3932 7 0.0157 0.3813 7 0.0092 0.3793 7 0.0174 0.3772 7 0.0129 0.3780 7 0.0130 0.3824 7 0.0168 0.3696 7 0.0179 0.3778 7 0.0144 0.3793 7 0.0124 0.3892 7 0.0292 0.3889 7 0.0160 0.3868 7 0.0112 0.3886 7 0.0136 0.3802 7 0.0167 0.3911 7 0.0093 0.3832 7 0.0158 0.3762 7 0.0162 0.3854 7 0.0146 0.3789 7 0.0154 0.3875 7 0.0132 0.3830 7 0.0131 0.3765 7 0.0124 0.3811 7 0.0096 0.3789 7 0.0110 0.3893 7 0.0164

207

Pb*/206 Pb* age (Ma)

%206 Pb*

Th/U

208

Pb*/206 Pb*

3816 7 27 3810 7 22 3850 7 14 3803 7 22 3859 7 16 3839 7 14 3788 7 35 3780 7 44 3824 7 16 3883 7 24 3837 7 15 3829 7 28 3820 7 21 3824 7 21 3841 7 27 3790 7 32 3823 7 23 3829 7 20 3868 7 45 3867 7 25 3858 7 18 3866 7 21 3833 7 27 3875 7 14 3845 7 25 3817 7 26 3853 7 23 3827 7 24 3861 7 21 3844 7 21 3818 7 20 3836 7 15 3827 7 18 3868 7 25

100 99.87 99.8 99.74 100 100 96.62 100 100 100 100 100 99.91 99.8 100 99.91 99.86 99.72 100 100 99.57 100 100 99.7 99.53 99.7 100 100 99.76 99.3 100 100 100 99.89

0.4427 0.5511 0.5682 0.5551 0.5435 0.5837 0.6046 0.7226 0.7054 0.6005 0.5421 0.6326 0.5583 0.6451 0.5519 0.6612 0.5895 0.6256 0.6506 0.6443 0.5988 0.67 0.6721 0.6594 0.6534 0.6198 0.5865 0.6133 0.7056 0.6575 0.731 0.6226 0.6641 0.6962

0.1367 0.1709 0.1856 0.1655 0.1481 0.1609 0.1735 0.2056 0.1954 0.2037 0.1912 0.1898 0.1744 0.2037 0.1953 0.1814 0.1916 0.1973 0.2049 0.1781 0.1752 0.2014 0.1853 0.1984 0.1881 0.2051 0.1822 0.1879 0.2059 0.1986 0.1896 0.1766 0.2166 0.2184

%206 Pb* is the percent of radiogenic 206 Pb. Isotopic ratios have been corrected for common Pb. U^Pb ages were determined by comparison with a working curve de¢ned by measurement of standard zircon AS-3 that yields concordant 206 Pb/238 U and 207 Pb/ 235 U ages of 1099.1 7 0.5 Ma by conventional methods [23]. Error demagni¢cation was applied to the data; all uncertainties are reported at the 1c level.

younger ages (Fig. 3). However, in spot analysis mode, the 2-D spatial selectivity of the ion microprobe is limited by the size of the primary beam, typically 10^30 Wm. In the case where a crystal contains symmetrical overgrowths, it is possible to use the ion microprobe in depth pro¢ling mode on an initially unpolished surface to increase spatial resolution by up to two orders of magnitude [24]. This approach, applied to zircons and described below, obtains a continuous age

pro¢le as well as provides geochemical signatures that could bear on the conditions of zircon growth coinciding with documented metamorphic episodes. The technique potentially transcends ambiguity of interpretations of zircon inheritance vs. protolith magmatic ages by linking the geochemistry of the host rock with the minute scale of growth and composition of an individual zircon at a spatial resolution unattainable in normal spot analysis of polished grains.

EPSL 6344 11-9-02

570

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

3.2. Zircon U^Th^Pb depth pro¢le analysis After evaluation by light microscopy, a separated euhedral zircon grain (GR9716_69) that yielded a concordant U^Pb age of V3.85 Ga [16] was plucked from its ion microprobe mount and an original crystal prism surface placed face down onto adhesive tape together with grains of standard zircon AS-3. The enclosing 1Q diameter mold was then recast with epoxy. When cured, the mount was removed from its adhesive backing, cleaned in distilled water and Au coated according to our usual procedures [25] but without ini-

tial polishing (Fig. 4). The internal distributions of U^Pb and Th/U in zircon GR9716_69 were then obtained by a high-resolution depth pro¢ling technique [24] utilizing a V5-nA primary O3 beam focused to a 30U20 Wm spot. Pre-sputtering times were 120 s and analysis comprised 15 cycles of 10-s measurements. Each data point listed in Table 2 is composed of one block of analyses made up of three cycles. Analytical sessions in depth pro¢ling mode involved a prolonged and continuous collection time (V30 min) from the same spot as the ion beam sputters into the unpolished zircon sample. The extended nature of this depth pro¢le required us to develop an approach di¡erent from that of conventional ion microprobe zircon geochronology where spots are analyzed on minerals mounted and polished to their centers. After sputtering a V5 Wm deep crater with the O3 ion beam, we measured the pit depth using a DekTakz surface pro¢lometer and then repolished by hand the sample surface using a 0.25 Wm diamond-impregnated lapping ¢lm under repeated supervision using re£ected-light microscopy of the grain until the crater was V1 Wm deep. By measuring the depth of the residual pit from the polishing, we could then accurately calculate the thickness of zircon removed from the polishing. Subsequently the zircon was reanalyzed in depth pro¢le mode by ion microprobe after cleaning and Au coating. We repeated this sequence twice to obtain a V16-Wm-long pro¢le from the surface of the grain to the center.

4. Results

Fig. 5. (a) U^Pb age vs. depth pro¢le in zircon GR9716_69. The spectrum of ages is interpreted to be due to a magmatic core with an age of 3828 7 8 Ma and metamorphic overgrowths at V3.6 Ga and 2.5^2.7 Ga. (b) Depth pro¢les of measured Th/U and 208 Pb/206 Pb. Note the correlation between Th/U increasing from V1032 to 0.8 and increasing age from V3.6 Ga to 3.83 Ga. Grey region shows predicted zircon Th/U crystallizing from a melt of Th/U = 3.8 assuming liq=zir KTh=U = 5 ( 7 factor of two) [30].

Our depth pro¢le data (Table 2 ; Fig. 5) reveal a concentrically zoned crystal containing the previously recognized age components in the Akilia orthogneisses [3,5^7,16,22,26,27]: between 0^3 Wm depth, concordant 207 Pb/206 Pb ages increase essentially monotonically from 2.5 to 2.7 Ga; ages then rise abruptly to a V3.6-Ga plateau between 3 and 5 Wm; between V8 and 16 Wm ages are consistent with a single population. The weighted mean of all 43 207 Pb/206 Pb ages between V8 and 16 Wm depth (Table 2) is 3828 7 8 Ma (2 c; MSWD = 1.2) (Fig. 5A).

EPSL 6344 11-9-02

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

Fig. 6. Correlation diagram of 208 Pb/206 Pb vs. Th/U for ion microprobe depth pro¢le data of zircon GR9716_69. The variability of data for the V3.6-Ga and V2.7-Ga zones of the zircon are in contrast to the relatively small spread of values for the 3830-Ga component.

4.1. Whole-rock and zircon Th/U chemistry These depth pro¢le data also show a clear correlation between both the measured Th/U and time-integrated Th/U (208 Pb/206 Pb) with apparent U^Pb age. Because zircon is the only magmatic phase in this rock that could signi¢cantly fractionate Th from U, the crystallizing tonalitic melt would maintain a Th/U ratio similar to the whole rock value of 3.8 (typical of continental crust [28]) until zircon became saturated (Table 1 ; Fig. 2) [29]. A zircon grown from a melt with Th/U = 3.8 is expected to have a Th/U zircon=melt = 0.2) [30]. The only ratio of V0.8 (i.e. DTh=U portion of zircon GR9716_69 characterized by a similar value is the 3.83-Ga component, which closely matches the predicted Th/U (Fig. 5B). In contrast, Th/U for the V3.6-Ga and V2.5^2.7Ga zones are indistinguishable from one another and are nearly two orders of magnitude lower than the 3.83-Ga portion (Fig. 6). Because there is agreement that zircon growth at ca. 2.7 Ga is metamorphic [3,5^7], the simplest explanation for the V3.6-Ga portion is that it too precipitated from an aqueous metamorphic £uid. During prograde metamorphism, Th/U generally increases in orthogneisses from an average crustal value of V4 to as high as 50, implying that U is

571

preferentially leached relative to Th and that metamorphic £uids are characterized by low Th/U [31,32]. Thus our expectation is that the Th/U ratio of zircon precipitated from a metamorphic £uid would be substantially lower than the magmatic value, both because of the low Th/U of the intergranular £uid and the preferential partitioning of U with respect to Th into the precipitating zircon. The combination of these two e¡ects could explain the decrease in Th/U by two orders of magnitude in the 2.5^2.7-Ga and V3.6-Ga overgrowths relative to the 3.83Ga core. We interpret these data to indicate that the only primary magmatic component of this zircon is associated with an age of 3828 7 8 Ma. 4.2. In situ measurements of zircon in petrographic thin sections We gained additional insight into the zircon age distribution by U^Pb dating zircons in petrographic thin section of GR9716. Zircons are the dominant Th/U carrying phase and occur in three textural relationships in GR9716, i.e. (1) small ( 6 50 Wm) zircons hosted entirely within minimum-melt leucosomes ; (2) similarly small zircons hosted by plagioclase ; and (3) zircons ranging in size from 50-Wm to rarely 250-Wm grains, hosted by ma¢c phases. Eighty-eight U^Pb ages from 71 grains were measured in situ by this technique. In general, ages span the range from ca. 3.8 to 2.5 Ga (Table 3 ; Fig. 7) with the exception of one zircon encompassed with in biotite that yielded an anomalous age of 4.1 Ga. Zircons in leucosomes yielded only V3.6-Ga ages. This latter observation is consistent with the earlier inference [20] that partial melting accompanied granulite facies metamorphism at that time [5] and potentially explains the variation of Th/U in V3.6-Ga zircons (Fig. 3). Both values consistent with primary magmas (V0.5) and those expected for precipitation from a metamorphic £uid ( 9 1031 ) are represented in our data (Fig. 7). We propose that the dissolution/re-precipitation of a signi¢cant fraction ( s 50%?) of the protolith zircon at V3.6 Ga via partial melting, solution in an aqueous £uid, or Ostwald ripening [33,34], provides a source of radiogenic Pb at that time

EPSL 6344 11-9-02

572

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

Table 3 Ion microprobe analyses of zircons in situ from orthogneiss sample GR9716 Grain spot

206

Pb*/238 U age

g1-1 g11-1 z11-1 z14-1 g6-1 g17-1 g17-2 z1-1 z1-2 z2-1 g4-2 g4-1 g16-1 g16-2 z13-1 g5-1 g2-1 z3-1 z4-1 z6-1 z8-1 z10-1 g3-2 g3-1 g9-1 g18-1 g14-1 z5-1 z7-1 g8-1 g12-1 z9-1 z12-1 z15-1 z16-1 z17-1 z18-1 15_2_29_1 15_2_29_2 15_2_33_1 15_2_38_1 15_2_38_2 15_2_39_1 15_2_40_1 15_2_42_1 15_2_5_1 15_2_6_1 16_2_10_1 16_2_11_1 16_2_1_1 16_2_21_1 16_2_23_1 16_2_24_1 16_2_2_1

3738 7 173 3444 7 105 3249 7 58 3452 7 49 3491 7 54 3475 7 62 3383 7 86 3295 7 69 2335 7 95 3773 7 64 3793 7 54 3376 7 47 3108 7 78 1690 7 40 1213 7 40 3801 7 96 2791 7 139 3614 7 71 3627 7 65 3349 7 82 3573 7 46 3201 7 88 3281 7 158 2702 7 65 1768 7 66 3083 7 81 2434 7 80 2811 7 77 3343 7 82 3129 7 99 3115 7 87 3464 7 43 3495 7 52 2935 7 62 3397 7 46 3174 7 33 3314 7 41 3577 7 47 3546 7 58 3550 7 34 3773 7 107 3730 7 68 3538 7 25 3313 7 18 2855 7 51 3316 7 29 2957 7 37 2817 7 18 3606 7 32 3693 7 21 3337 7 37 3197 7 47 3312 7 29 3389 7 26

207

Pb*/235 U age

3963 7 66 3440 7 40 3496 7 21 3529 7 18 3631 7 20 3617 7 22 3570 7 33 3472 7 26 3002 7 46 3746 7 22 3755 7 18 3554 7 17 3438 7 31 2706 7 27 2368 7 34 3756 7 33 3053 7 60 3624 7 25 3606 7 25 3514 7 30 3612 7 16 3222 7 35 3480 7 60 2896 7 31 2850 7 44 3428 7 34 2556 7 38 3002 7 34 3465 7 32 3468 7 40 3333 7 36 3623 7 16 3625 7 20 3307 7 27 3517 7 17 3413 7 14 3545 7 15 3606 7 24 3535 7 42 3569 7 17 3597 7 45 3650 7 38 3545 7 10 3457 7 8 3021 7 39 3457 7 17 3072 7 32 2840 7 14 3611 7 17 3699 7 10 3379 7 18 3303 7 24 3417 7 18 3513 7 12

EPSL 6344 11-9-02

207

Pb*/206 Pb* age

4079 7 18 3437 7 14 3641 7 4 3574 7 6 3709 7 6 3696 7 8 3677 7 10 3576 7 7 3485 7 19 3732 7 7 3735 7 5 3656 7 6 3636 7 4 3586 7 14 3598 7 14 3732 7 11 3231 7 12 3629 7 7 3594 7 7 3610 7 10 3633 7 5 3236 7 14 3597 7 5 3033 7 20 3741 7 14 3636 7 11 2654 7 16 3133 7 15 3536 7 8 3670 7 5 3466 7 13 3713 7 3 3697 7 7 3540 7 9 3587 7 6 3556 7 7 3678 7 4 3623 7 14 3528 7 28 3580 7 10 3500 7 20 3607 7 24 3548 7 4 3542 7 6 3133 7 28 3540 7 11 3148 7 25 2857 7 11 3614 7 11 3702 7 7 3404 7 11 3369 7 13 3478 7 11 3584 7 5

Th/U 0.0740 0.1480 0.0700 0.0910 0.4070 0.3640 0.6030 0.3960 0.4480 0.4560 0.7630 0.7570 0.3320 0.7210 0.3610 0.4050 0.1040 0.2250 0.9690 0.5920 0.1720 0.4660 0.0930 0.1470 0.2780 0.3310 0.0440 0.2760 0.1490 0.6620 0.1190 0.4430 0.0930 0.1560 0.1630 0.2310 0.3080 0.2160 0.2660 0.5920 0.1570 0.2360 0.1200 0.0778 0.2520 0.2150 0.0808 0.0848 0.2190 0.3950 0.3140 0.1060 0.0802 0.1210

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

573

Table 3 (Continued). Grain spot

206

Pb*/238 U age

16_2_35_1 16_2_43_1 16_2_47_1 16_2_4_1 16_2_6_1 16_2_7_1 15_10_1_1 15_10_1_2 15_10_1_3 15_10_3_1 15_10_3_3 15_6_1_1 15_6_2_2 15_6_3_1 15_6_4_1 15_6_5_1 15_6_6_1 15_6_7_1 15_7_10_1 15_7_11_1 15_7_11_2 15_7_1_1 15_7_2_1 15_7_2_2 15_7_2_3 15_7_3_1 15_7_4_1 15_7_5_1 15_7_6_1 15_7_7_1 15_7_7_2 15_7_8_1 15_7_9_1 15_7_9_2

2729 7 18 2826 7 22 3714 7 42 3170 7 30 3571 7 42 2523 7 18 3890 7 211 4078 7 265 3432 7 224 3222 7 251 3174 7 160 2893 7 270 2587 7 63 2569 7 89 3524 7 127 3656 7 77 2621 7 93 3266 7 121 3374 7 198 3395 7 218 2985 7 129 3308 7 160 3477 7 128 3574 7 176 3204 7 190 3550 7 81 3196 7 100 3225 7 97 3060 7 145 3065 7 120 3859 7 267 2778 7 148 2905 7 130 2740 7 123

207

Pb*/235 U age

2755 7 25 2915 7 13 3619 7 21 3270 7 15 3578 7 21 2571 7 11 3715 7 74 3833 7 91 3494 7 87 3314 7 105 3381 7 72 2741 7 111 2601 7 29 3055 7 41 3516 7 47 3611 7 28 3061 7 43 3257 7 48 3429 7 76 3386 7 84 3227 7 56 3444 7 64 3555 7 48 3530 7 64 3349 7 75 3554 7 30 3356 7 40 3399 7 40 3322 7 58 3341 7 51 3612 7 93 3025 7 67 3245 7 56 2956 7 56

207

Pb*/206 Pb* age

2775 7 23 2977 7 10 3567 7 13 3332 7 8 3582 7 12 2609 7 8 3622 7 17 3707 7 23 3529 7 20 3370 7 28 3507 7 25 2631 7 13 2612 7 8 3392 7 8 3512 7 6 3586 7 5 3364 7 13 3251 7 12 3461 7 12 3380 7 22 3381 7 21 3524 7 11 3599 7 7 3505 7 13 3437 7 16 3557 7 4 3454 7 12 3502 7 9 3484 7 11 3511 7 13 3478 7 12 3193 7 22 3461 7 14 3106 7 21

Th/U 0.0166 0.0436 0.3480 0.1680 0.4350 0.0123 0.9770 0.2770 0.8750 0.1840 0.9940 0.0767 0.0539 0.0765 0.1230 0.1690 0.2340 0.1540 0.1220 0.3850 0.1590 0.6300 0.6490 0.2740 0.1120 0.0688 0.1510 0.1070 0.3080 0.6420 0.1090 0.2840 1.3700 0.0888

Ages are expressed in millions of yr (Ma). Nomenclature is sample+grain number+ion microprobe spot number. ‘g’, ‘z’, ‘15’ and ‘16’ analyses were done during separate sessions; ‘15_’ and ‘16_’ are aliquots of the same sample GR9716. Isotopic ratios have been corrected for common Pb. U^Pb ages were determined by comparison with a working curve de¢ned by measurement of standard zircon AS-3 that yields concordant 206 Pb/238 U and 207 Pb/235 U ages of 1099.1 7 0.5 Ma by conventional methods [23]. Error demagni¢cation was applied to the data; all uncertainties are reported at the 1c level.

to reset the common Pb isotopic composition of the coexisting feldspars. An anomalous V4.1-Ga zircon age we observed for one grain in thin section is unusual in both its surprising antiquity (no U^Pb zircon age this old has been reported for the ‘Itsaq Gneiss Complex’) and low Th/U value (0.074) relative to the cluster of ages at V3.8 Ga (Table 3). While we noted earlier the unlikelihood of the parent tonalite melt containing a high fraction of inher-

ited zircons (see [6]), this speci¢c and isolated occurrence can be explained by the fact that the zircon is included within biotite. This association suggests that zircon survived the magmatic episode because it was incorporated into the magma late in its crystallization history and was temporarily armored against dissolution by the encompassing stable phase. The apparent rarity of this relationship suggests that the postulated armoring mechanism is not robust and thus the inherited

EPSL 6344 11-9-02

574

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

Fig. 7. Results of zircon U^Th^Pb in situ studies of orthogneiss GR9716 plotted with data from [3,6,16] and listed on Table 3.

zircon is unlikely to have been transported far from its source, otherwise the grain and its host biotite would have been resorbed by the tonalite. Indeed, based on this ¢nding it is of interest whether there may be rocks even older than the 3.83-Ga Akilia orthogneiss present somewhere in the ‘Itsaq Gneiss Complex’ of West Greenland and perhaps in the vicinity of Akilia.

5. Conclusions Our conclusions regarding the age and origin of the contested orthogneiss on Akilia are as follows. (1) Age pro¢les from near surface ( 9 5 Wm) regions of individual zircons can be obtained using the depth pro¢ling mode of the ion microprobe. The increased spatial resolution of this zircon geochronological technique permits detection of age variations at the V200-nm scale. By employing sequential repolishing, depth^age-composition pro¢les of zircons of essentially unlimited length can be obtained. (2) U^Pb age depth pro¢ling of a complex tonalitic orthogneiss from Akilia, West Greenland, reveals evidence for three phases of concentric zircon growth at 3.83 Ga, V3.6 Ma and 2.7^ 2.5 Ga.

(3) Zircon growth at both V3.6 Ga and 2.7^ 2.5 Ga is consistent with precipitation from a metamorphic £uid. Only the zircon core is consistent with crystallization from magma of the composition of the host rock. (4) The crystallization age of the previously contested Southwest Akilia tonalitic orthogneiss is 3828 7 8 Ma. Hence, sediments [3] and evidence for life therein [2] on Akilia are s 3.83 Ga in age. (5) That these data do not agree with commonPb compositions of the Akilia orthogneiss [12] calls for quantitative studies of open-system behavior of Pb during multiple periods of highgrade (granulite) metamorphism. An important aspect of future work would be the evaluation of £uid transport of U and Th that proceeded during granulitization of, for example, West Greenland orthogneisses. (6) An anomalously old (V4.1 Ga) zircon (Table 3; g1-1) measured in situ in thin section may point to the unforeseen existence of primordial crust in the vicinity of the ‘Itsaq Gneiss Complex’.

Acknowledgements We are grateful for technical assistance by C.D. Coath, G. Jarzebinski and E. Catlos. This work was supported by grants from NSF Life in Extreme Environments (LExEn) program EAR9978241 (SJM and TMH), an NSF postdoctoral fellowship from grant EAR-9704651 (SJM), the NASA Astrobiology Institute and DoE. Constructive comments to the manuscript by Bill Gri⁄n, Minik Rosing and Je¡ Vervoort and editorial handling by Bernie Wood are appreciated. Discussions with A.P. Nutman and C.R.L. Friend regarding the ¢eld relationships have been valuable. We acknowledge facility support from the Instrumentation and Facilities Program of the National Science Foundation.[BW] References [1] J.R. Ridley, J.R. Vearncombe, H.A. Jelsma, Relations between greenstone belts and associated granitoids, in: M. deWit, L.D. Ashwal (Eds.), Greenstone Belts, Oxford

EPSL 6344 11-9-02

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9] [10] [11]

[12]

[13]

[14]

[15]

[16]

Monographs on Geology and Geophysics 35, Clarendon Press, Oxford, 1997, pp. 376^397. S.J. Mojzsis, G. Arrhenius, K.D. McKeegan, T.M. Harrison, A.P. Nutman, C.R.L. Friend, Evidence for life on Earth before 3,800 million years ago, Nature 384 (1996) 55^59. A.P. Nutman, S.J. Mojzsis, C.R.L. Friend, Recognition of v 3850 Ma water-lain sediments in West Greenland and their signi¢cance for the early Archaean Earth, Geochim. Cosmochim. Acta 61 (1997) 2475^2484. V.R. McGregor, B. Mason, Petrogenesis and geochemistry of metabasaltic and metasedimentary enclaves in the Amitsoq gneisses, West Greenland, Am. Min. 62 (1977) 887^904. A.P. Nutman, V.R. McGregor, C.R.L. Friend, V.C. Bennett, The Itsaq Gneiss Complex of southern West Greenland; the world’s most extensive record of early crustal evolution (3,900^3,600 Ma), Precambr. Res. 78 (1996) 1^ 39. M. Whitehouse, B.S. Kamber, S. Moorbath, Age signi¢cance of U^Th^Pb zircon data from early Archaean rocks of West Greenland ^ a reassessment based on combined ion-microprobe and imaging studies, Chem. Geol. 160 (1999) 201^224. M. Whitehouse, Time constraints on when life began: The oldest record of life on Earth?, Geochem. News 103 (2000) 10^14. M. Schidlowski, A 3,800-million-year isotopic record of life from carbon in sedimentary rocks, Nature 333 (1988) 313^318. J.M. Hayes, The earliest memories of life on Earth, Nature 384 (1996) 21^22. H.D. Holland, Evidence for life on Earth more than 3850 million years ago, Science 275 (1997) 38^39. M. Rosing, 13 C-depleted carbon microparticles in s 3700 Ma sea£oor sedimentary rocks from West Greenland, Science 283 (1999) 674^676. S. Moorbath, M.J. Whitehouse, B.S. Kamber, Extreme Nd-isotope heterogeneity in the early Archaean ^ fact or ¢ction? Case histories from northern Canada and West Greenland, Chem. Geol. 135 (1997) 213^231. B.S. Kamber, S. Moorbath, Initial Pb of the Am|“tsoq gneiss revisited: Implication for the timing of the early Archaean crustal evolution in West Greenland, Chem. Geol. 150 (1998) 19^41. A.D. Anbar, K.J. Zahnle, G. Arnold, S.J. Mojzsis, Extraterrestrial iridium, sediment accumulation and the habitability of the early Earth’s surface, J. Geophys. Res. 106 (2001) 3219^3236. S.J. Mojzsis, G. Ryder, Extraterrestrial accretion to the Earth and Moon ca. 3.85 Ga, in: B. Peuckner-Ehrenbrink, B. Schmitz (Eds.), Accretion of Extraterrestrial Matter Throughout Earth History, Kluwer, Amsterdam, 2001, pp. 423^446. S.J. Mojzsis, T.M. Harrison, Vestiges of a beginning: Clues to the emergent biosphere recorded in the oldest known sedimentary rocks, GSA Today 10 (2000) 1^6.

575

[17] B.M. Jakosky, R.J. Phillips, Mars’ volatile and climate history, Nature 412 (2001) 237^244. [18] T.M. Harrison, E.B. Watson, Kinetics of zircon dissolution and zirconium di¡usion in granitic melts of variable water content, Contrib. Min. Pet. 84 (1983) 66^72. [19] P.J. Wyllie, M.B. Wolf, S.R. van der Laan, Conditions for the formation of tonalites and trondjemites: Magma sources and products, in: M.J. de Wit, L.D. Ashwal (Eds.), Greenstone Belts, Oxford Univ. Press, Oxford, 1997, pp. 256^265. [20] W.L. Gri⁄n, V.R. McGregor, A.P. Nutman, P.N. Taylor, D. Bridgwater, Early Archaean granulite facies metamorphism south of Ameralik, West Greenland, Earth Planet. Sci. Lett. 50 (1980) 59^74. [21] D.J. Cherniak, Di¡usion of Pb in feldspar measured by Rutherford backscattering spectroscopy, EOS Trans. Am. Geophys. Union 73 (1992) 641. [22] V.R. McGregor, Initial Pb of the Am|“tsoq gneiss revisited: Implications for the timing of Early Archaean crustal evolution in West Greenland, Comment. Chem. Geol. 166 (2000) 301^308. [23] J.B. Paces, J.D. Miller, Precise U^Pb ages of Duluth Complex and related ma¢c intrusions, northeastern Minnesota ^ Geochronological insights to physical, petrogenetic, paleomagnetic and tectonomagmatic processes associated with the 1.1 Ga midcontinent rift system, J. Geophys. Res. 98 (1993) 13997^14013. [24] M. Grove, T.M. Harrison, Monazite Th^Pb age depth pro¢ling, Geology 27 (1999) 487^490. [25] X. Quidelleur, M. Grove, O.M. Lovera, T.M. Harrison, A. Yin, F.J. Ryerson, Thermal evolution and slip history of the Renbu Zedong Thrust, southeastern Tibet, J. Geophys. Res. 102 (1997) 2659^2679. [26] P.D. Kinny, 3820 Ma zircons from a tonalitic Am|“tsoq gneiss in the Godthafib district of southern West Greenland, Earth Planet. Sci. Lett. 79 (1986) 337^ 347. [27] A.P. Nutman, V.C. Bennett, C.R.L. Friend, V.R. McGregor, The Early Archaean Itsaq Gneiss Complex of southern West Greenland: The importance of ¢eld observations in interpreting age and isotopic constraints for early terrestrial evolution, Geochim. Cosmochim. Acta 64 (2000) 3035^3060. [28] S.R. Taylor, S.M. McLennan, The Continental Crust: Its Composition and Evolution, Blackwell Scienti¢c, Oxford, 1985. [29] J.C. Ayers, E.B. Watson, Solubility of apatite, monazite, zircon, and rutile in supercritical £uids with implications for subduction zone geochemistry, Philos. Trans. R. Soc. A335 (1991) 365^375. [30] G. Mahood, W. Hildreth, Large partition coe⁄cients for trace-elements in high-silica rhyolites, Geochim. Cosmochim. Acta 47 (1983) 11^30. [31] A.D. Nozhkin, O.M. Turkina, Radiogeochemistry of the charnokite^granulite complex, Sharyzhalgay Window, Siberian Platform, Geochem. Int. 32 (1995) 62^78. [32] H.R. Rollinson, B.F. Windley, Selective elemental deple-

EPSL 6344 11-9-02

576

S.J. Mojzsis, T.M. Harrison / Earth and Planetary Science Letters 202 (2002) 563^576

tion during metamorphism of Archean granulites, Contrib. Mineral. Petrol. 72 (1980) 257^263. [33] R.L. Joesten, Kinetics of coarsening and di¡usion-controlled mineral growth, Rev. Mineral. 26 (1991) 507^582. [34] P.W.O. Hoskin, L.P. Black, Metamorphic zircon forma-

tion by solid-state recrystallization of protolith igneous zircon, J. Metamorph. Geol. 18, pp. 423^439. [35] C.M. Fedo, M.J. Whitehouse, Metasomatic origin of quartz-pyroxene rock, Akilia, Greenland, and implications for Earth’s earliest life. Science 296, pp. 1448^1452.

EPSL 6344 11-9-02